Safety and immunogenicity of a new13-valent pneumococcal conjugatevaccine versus a licensed 7-valentpneumococcal conjugate vaccine:a study protocol of a randomisednon-inferiority trial in China

Jing Jing Chen,1,2 Lin Yuan,2 Zhen Huang,2 Nian Min Shi,3 Yu Liang Zhao,4

Sheng Li Xia,5 Guo Hua Li,6 Rong Cheng Li,7 Yan Ping Li,7 Shu Yuan Yang,2

Jie Lai Xia1

To cite: Chen JJ, Yuan L,Huang Z, et al. Safety andimmunogenicity of a new13-valent pneumococcalconjugate vaccine versus alicensed 7-valentpneumococcal conjugatevaccine: a study protocol of arandomised non-inferioritytrial in China. BMJ Open2016;6:e012488.doi:10.1136/bmjopen-2016-012488

▸ Prepublication history forthis paper is available online.To view these files pleasevisit the journal online(http://dx.doi.org/10.1136/bmjopen-2016-012488).

Received 1 May 2016Revised 3 September 2016Accepted 8 September 2016

For numbered affiliations seeend of article.

Correspondence toProfessor Jie Lai Xia;jielaixia@yahoo.com

ABSTRACTIntroduction: The invasive pneumococcal diseases(IPDs) caused by Streptococcus pneumoniae pose anenormous threat to children under 5 years of age.However, routine use of pneumococcal conjugatevaccines could aid in reducing the incidence of IPDs.The purpose of this clinical trial is to assess the non-inferiority of the investigational 13-valentpneumococcal conjugate vaccine (PCV13) to thecurrently licensed 7-valent pneumococcal conjugatevaccine (PCV7).Methods and analysis: 1040 infants will receive athree-dose series of either PCV13 or PCV7 at ages 3, 4and 5 months, respectively, and a booster dose at12–15 months. Primary end points are the percentageof participants reaching a serotype-specific IgGconcentration of ≥0.35 µg/mL and the IgG antibodygeometric mean concentrations (GMCs) measured30 days after the primary immunisation. Secondary endpoints include the percentage of vaccine recipientsreaching a serotype-specific IgG concentrationthreshold of 1.0 µg/mL, the percentage of participantsreaching the pneumococcal opsonophagocytic assay(OPA) titre threshold of 1:8, and the geometric meantitres (GMTs) of OPA measured 30 days after primaryand booster doses. The number of standard IgGresponders and IgG GMCs measured 30 days after thebooster immunisation will also be determined. Toevaluate differences between two groups, thesequential testing of the non-inferiority of PCV13 forthe seven common serotypes and its effectiveness intreating the six additional serotypes will be performed.Ethics and dissemination: Ethics approvals havebeen granted by the Ethics Committees at the threeprovinces involved in this study: Shanxi, Henan andHebei. The trial will be reported in accordance with theCONSORT guidance.Trial registration number: NCT02736240.

INTRODUCTIONBackgroundStreptococcus pneumoniae (S. pneumoniae orpneumococcus) is a leading cause of bacterialpneumonia, meningitis and sepsis in chil-dren, and thus poses a major threat to publichealth around the world. According to aWHO investigation in 2005, an estimated 0.7to 1 million children under the age of5 years die from pneumonia every year, mostof whom lived in developing countries.1 Thetop 10 countries with the highest number ofreported pneumococcal cases were all inAsia and Africa, with China accounting for12% of the total.2 Immunisation against Hib,

Strengths and limitations of this study

▪ This is the first study to investigate the immuneresponses induced by a locally developed andmanufactured 13-valent pneumococcal conjugatevaccine (PCV13) among Chinese infants.

▪ A total of 1040 infants will be recruited tosupport the randomised non-inferiority studydesign and a sequential test method is adopted.

▪ Primary outcomes are proportions of vaccinerecipients achieving a predefined specific-serotype IgG concentration threshold and geo-metric mean IgG antibody concentrations mea-sured 30 days after primary immunisation, bothof which are recommended by the WHO.

▪ Besides the complete evaluation of immunogen-icity and safety of the new PCV13, not involvingthe assessment of concomitant immunisationwith other vaccines specified in paediatric vac-cination might be seen as a potential limitationof this trial.

Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488 1

Open Access Protocol

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

pneumococcus, measles and pertussis is the most effect-ive way to prevent pneumonia.3 Routine use of the7-valent pneumococcal conjugate vaccine (PCV7) hasresulted in a dramatic reduction in the incidence ofinvasive pneumococcal disease (IPD) attributable topneumococcal serotypes contained in the vaccine.4

However, there is still an urgent need for a pneumococ-cal conjugate vaccine with expanded coverage for IPDresulting from serotypes not included in PCV7.5

On 24 February 2010, the US Food and DrugAdministration licensed a 13-valent pneumococcal con-jugate vaccine (PCV13) for children from 2 months to5 years of age. Since then, several studies have been con-ducted on PCV13 in many countries except China. Thelicensed PCV13 has been shown to be as effective asPCV7 in the prevention of disease caused by the sevencommon serotypes and could provide expanded protec-tion against the six additional serotypes with a compar-able safety profile.6 When coadministered withdiphtheria, tetanus and acellular pertussis vaccine(DTaP), PCV13 is well tolerated and as immunogenic asPCV7.7 The flexible primary immunisation schedules ofPCV13 for infants found no significant differences inantibody levels for almost all serotypes after the boosterdose at 12 months of age8 9 and is thus recommendedby the WHO.10 Furthermore, the clinical efficacy ofPCV13 for multiple purposes in age groups rangingfrom 6 to over 70 years of age has been confirmed.11–16

Nevertheless, it is still unknown whether the licensedPCV13 or a new PCV13 will be effective for the Chinesepopulation until now.To address the issue, we plan to initiate this trial for a

prelicensing evaluation of a new PCV13. The investiga-tional PCV13 (batch number 201511003) is manufac-tured by Yuxi Walvax Biotechnology Co (an affiliate ofWalvax Biotechnology Co) and supplied in penicillinbottles. Each 0.5 mL dose contains 2.2 µg pneumococcalpolysaccharide serotypes 1, 3, 4, 5, 6A, 7F, 9V, 14, 18C,19A, 19F and 23F and 4.4 µg pneumococcal polysacchar-ide serotype 6B; each serotype is conjugated to a tetanustoxoid carrier protein and adsorbed on aluminiumphosphate. In accordance with the guidelines formu-lated by the WHO,17 and the requirements establishedby national regulatory authorities, licensing a newvaccine requires randomised clinical trials to show thenon-inferiority of the new vaccine to an existingpneumococcal conjugate vaccine. PCV7 will serve as thecontrol vaccine, since it is the only pneumococcal conju-gate vaccine that has been licensed in mainland China.The control PCV7 (Prevnar, batch numberH29315H13880) is manufactured by Wyeth and suppliedin prefilled 0.5 mL syringes. Each 0.5 mL dose contains2 µg pneumococcal polysaccharide serotypes 4, 9V, 14,18C, 19F and 23F and 4 µg pneumococcal polysacchar-ide serotype 6B; each serotype is conjugated to aCRM197 carrier protein and adsorbed on 0.5 mg alu-minium phosphate.

ObjectivesThe purpose of this study is to (1) determine if theseven common serotype-specific pneumococcal IgGresponses elicited by PCV13 are all non-inferior to thoseelicited by PCV7 and (2) determine if the six additionalserotype-specific pneumococcal IgG responses inducedby PCV13 are all clinically effective 30 days after theprimary immunisation.We hypothesise that for each of the seven common

pneumococcal serotypes (4, 6B, 9V, 14, 18C, 19F, 23F)(1) the lower bound of the 97.5% (98.75%, one-sided)CI for the difference (PCV13—PCV7) in the proportionof vaccine recipients reaching the serotype-specific IgGconcentration threshold of 0.35 μg/mL is no < −10%; or(2) The lower bound of the 97.5% (98.75%, one-sided)CI of the serotype-specific IgG geometric mean concen-tration (GMC) ratio (PCV13/PCV7) is > 0.5 (−0.3, loga-rithmically transformed by base 10).As a parallel hypothesis, we expect that for each of the

six additional pneumococcal serotypes (1, 3, 5, 6A, 7F,19A) (1) The proportion of the serotype-specific IgGresponders achieving an IgG threshold of 0.35 μg/mL inthe PCV13 group is greater than that of the PCV7group; and (2) the lower limit of the 95% (2-sided) CIfor the proportion of vaccine recipients reaching aserotype-specific IgG concentration threshold of0.35 μg/mL is ≥70%.

Trial designThis study is a multicentre phase III randomised parallelgroup, allocation concealed non-inferiority trial with twogroups to evaluate the immunogenicity and safety ofPCV13 compared with PCV7 among healthy infants inChina.

METHODSStudy settingThe study includes six sites scattered in three provincesin North China. Each site has been ratified by theCenter for Food and Drug Inspection, a department ofChina Food and Drug Administration, to join theresearch team of this trial and recruit participants.

Eligibility criteriaInclusion criteriaParticipants will only be recruited when he or she meetsall of the following criteria at the time of randomisation:1. Healthy infants, as established by medical history and

clinical examination before entering into the study,who are 3 months old at the time of the firstvaccination

2. Obtained written informed consent from the legallyresponsible representative(s)

3. Have the compliance of the participants’ legallyresponsible representative(s) with the requirementsof the protocol

2 Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

4. No pneumococcal vaccine history or use of other pre-ventive vaccine at least 7 days before study entry

5. Axillary temperature ≤37°CIn addition, participants must not encounter any of thefollowing exclusion criteria:1. Vaccinated by on sale or other on trial pneumococ-

cal vaccines2. Have had an invasive disease with S. pneumoniae as

the verified cause3. Severely allergic to any vaccines or drugs, or sensi-

tive to temperatures ≥39°C associated with bio-logical products inoculation

4. Birth weight <2.5 kg5. History or family history of convulsion, seizure,

encephalopathy or neurological disorder6. Born from abnormal labours (difficult labours,

aided by instruments) or have a history of asphyxiaor nerve damage

7. Confirmed thrombocytopenia or a history of bloodcoagulation disorder

8. Confirmed pathological jaundice9. Any confirmed or suspected immunodeficiency

involving immunosuppressive therapy (radiotherapy,chemotherapy, corticosteroid hormone, antimetab-olite, cytotoxic drugs), HIV infection, etc

10. Any confirmed or suspected congenital defect orserious chronic illnesses (trisomy 21 syndrome, dia-betes, sickle cell anaemia, neurological disorder orGuillain-Barré syndrome)

11. Any confirmed or suspected diseases includingrespiratory diseases, acute infection, active period ofchronic disease, cardiovascular disease, hepatic-nephrotic disease, cancer and dermatosis

12. Administration of immunoglobulin and/or anyblood products, except hepatitis B hyperimmuneglobulin (HBlg)

13. Concurrent participation in another clinical study14. Other exclusion criteria affirmed by investigators

InterventionEligible infants will be randomised to either the PCV13or the PCV7 group to receive a three-dose series ofPCV13 or PCV7 at ages 3, 4 and 5 months, respectivelyand a booster dose between 12 and 15 months. For allparticipants, blood samples will be obtained at four dif-ferent time points: immediately before the first dose,30 days after the primary series, immediately before thebooster dose, and 30 days after the toddler dose.Serum concentrations of anticapsular polysaccharide

IgG for each of the 13 pneumococcal serotypes will bemeasured for all participants at the previously men-tioned time points, using the standardised ELISAmethod.18 Additionally, serum functional opsonophago-cytic activity (OPA) for each serotype will be assayed atthe last 3 time points for a randomly selected subset of∼100 participants in each group (200 infants in total),composed of infants assigned the first 20–36 participantnumbers from each of the six sites.

Each participant will be observed for 30 min aftereach vaccination for any immediate reaction by theresearch staff. Any noted adverse events (AEs) will berecorded at that time. After leaving the site, local reac-tions (pain, redness, swelling and pruritus), systemicevents (somnolence, vomiting, diarrhoea, crying andirritability), axillary temperature, other AEs and con-comitant medications to treat or prevent symptoms willbe recorded in daily cards by the parents/legally respon-sible representatives for 8 days. On the eighth day aftervaccination, the documents will be returned to the site.AEs occurring between the 8th and the 30th day aftereach dose together with the usage of relevant concomi-tant medications will be recorded in connection cardsand handed in at the next visit. Serious adverse events(SAEs) will be collected throughout the study until6 months after the last dose. The intervention proce-dures are shown in table 1.

Outcome measuresPrimary immunogenicity outcomesFor each of the 13 pneumococcal serotypes covered inPCV13:1. Proportion of vaccine recipients reaching the

serotype-specific IgG concentration threshold of0.35 μg/mL 30 days after the primary series.

2. GMC of serotype-specific IgG antibodies 30 days afterthe primary series.

Secondary immunogenicity outcomesFor each of the 13 pneumococcal serotypes covered inPCV13:1. Proportion of vaccine recipients reaching the OPA

threshold of 1:8 30 days after the primary series.2. Geometric mean titre (GMT) of OPA 30 days after

the primary series.3. Proportion of vaccine recipients reaching the

serotype-specific IgG concentration threshold of1.0 μg/mL 30 days after the primary series.

4. Proportion of vaccine recipients reaching theserotype-specific IgG concentration threshold of0.35 μg/mL 30 days after the booster dose.

5. GMC of serotype-specific IgG antibodies 30 days afterthe booster dose.

6. Proportion of vaccine recipients reaching an OPAthreshold of 1:8 30 days after the booster dose.

7. GMT of OPA 30 days after the booster dose.8. Proportion of vaccine recipients reaching a serotype-

specific IgG concentration threshold of 1.0 μg/mL30 days after the booster dose.

Safety outcomesSafety outcomes include all recorded local reactions andsystemic AEs that occur within the 30 days following theprimary series and booster doses and the percentage ofvaccine recipients that reported each AE.

Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488 3

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

Trial durationThe study will last for 26 months. If unsuspected acuteabnormal reactions occur during the trial, or the dailyproportion of participants who experienced AEs ratedgrade 3 or above reaches 15% after any dose vaccination,the trial should be suspended. A panel will be involvedin discussions of the safety profile and decide whetherthe study should be continued.

Sample sizeFor each of the seven common serotypes in PCV13 andPCV7: On the basis of data from previous studies, weassume that the percentage of vaccine recipients reach-ing the serotype-specific IgG concentration threshold of0.35 μg/mL will be 85%.19 The non-inferior margin forthe trial will be set at 10%. If the non-inferiority doesexist for a proportion of the test group, a sample size of434 infants per group, 868 total, will provide 80% powerto ensure that the lower limit of 97.5% two-sided CIfor difference (test group–control group) will be no< −10%. We also postulate that the common variance ofserotype-specific IgG GMC, which has been logarithmic-ally transformed by base 10, will not be >1.0. A samplesize of 379 infants per group, 758 total, will have 80%power to detect whether the lower limit of 97.5% two-sided CI for the serotype-specific IgG GMC ratio (testgroup to control group) is above 0.5 or not. To takeaccount of α inflation and β consumption, Bonferronicorrection will be employed to α, producing correctedα=0.025/2=0.0125 and corrected power=1–0.2/7=0.97.For each of the six additional serotypes in PCV13 only:

On the basis of the preclinical research, we hypothesisethat the proportion of vaccine recipients reaching theserotype-specific IgG concentration threshold of

0.35 μg/mL in the test group will be about 80% whilethat in the control group will not be >50%. If there is anotable difference between the two groups, a samplesize of 274 infants per group, 548 total, will provide 80%power to demonstrate that the lower limit of 95% two-sided CI for a proportion of the test group will be no<70% (the target rate). To control for family-wise type IIerror, power will be adjusted to 0.97 (equals to 1–0.2/6).Final decision: Theoretically, a total of 868 evaluable

participants (434 per group) should enable the trial toachieve the primary outcomes. To take blood draw fail-ures and blood collection refusals into consideration, wewill allow for an anticipated 20% dropout rate.20

Therefore, the total sample size required is 1040 infants(520 per group).

Randomisation and allocationThe allocation sequence is generated by theDepartment of Health Statistics of Fourth MilitaryMedical University using the SAS (SAS Institute , Cary,North Carolina, USA) Plan procedure. In order to guar-antee the group balance in different sites, the block ran-domisation kit will be used. Each site will receiveconsecutively coded vaccines.Owing to the difference in appearance of PCV13 and

PCV7, the administrators cannot be blinded, but thegroup assigners, laboratory staff and investigators will be.Before recruitment, all vaccines will be masked in identi-cal white small boxes and labelled with vaccine numbersbased on the allocation schedule by those who are notinvolved in the clinical trial procedures. Each vaccinenumber is composed of two parts: the subject identifica-tion number and the dose specification number. Fourvaccines with the same subject number prefix will

Table 1 The intervention procedure

Primary series Booster dose

Interventions 3 months 4 months 5 months

30 days

postdose

12–15

months

30 days

postdose

6 months

postdose

Informed consent X

Eligibility screen X

Randomisation X

Temperature X X X X

Blood draw X X X X

Vaccination X X X X

Safety observations*

Immediate reaction ◎ ◎ ◎ ◎Local reactions and

systemic events

▲ ▲ ▲ ▲

Adverse events ◆ ◆ ◆ ◆Serious adverse

events

★ ★ ★ ★ ★ ★ ★

*Safety observations occur after vaccination for a planned period.◎Immediate reaction will be observed for 30 min after vaccination.▲Local reactions and systemic events will be actively collected for 8 days after vaccination.◆Other adverse events will be recorded by parents/legally responsible representative(s) for 30 days after vaccination.★Serious adverse events will be self-reported by parents/legally responsible representative(s) from the first visit until the 6 months after thelast dose.

4 Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

belong to the same participant. In case any unexpectedsituations arise, additional vaccines including PCV13and PCV7 in a ratio of 2:1 will be prepared and wrappedseparately in individual white boxes labelled withnumbers from B0001 to B0540. The link between theadditional vaccine number and subject number will beestablished based on an allocation sequence list byMedidata Rave—the electronic data capture (EDC)system for this study. Emergency letters will be sent toeach of the sites and properly preserved until needed.Each eligible participant will be sequentially allotted a

unique subject number according to the site allocationlist by authorised assigners, and thus will have an equalchance of being assigned to either the test group or thecontrol group. Blood samples will be labelled withsubject numbers and collected independently of the vac-cination process. The inoculator is responsible fortaking out the vaccine related to a specific subjectnumber and administering the injection. Inoculatorsmust not share any allocation information with investiga-tors or those involved in follow-up procedures.

WithdrawalThe participants’ parents or legally responsible represen-tatives will be aware that their decision to take part inthe study is voluntary and that they are free to withdrawconsent at any time. All withdrawals from the rando-mised intervention will be reported. The investigatormay drop participants during the study in the case ofintercurrent illness, AEs, SAEs, protocol violations orother reasons that meet the exclusion criteria.

Data managementThe trial data will be managed using the Medidata RaveEDC system. The database software provides several fea-tures to help maintain data quality. These featuresinclude, but are not limited to, keeping audit trails onall data fields, allowing custom functions in configura-tions and opening both automatic and manual queries.At the same time, the original record cards will bedesigned in accordance with the protocol to collect rawclinical data and to support the trial procedures. Thesource data obtained from each participant will beexamined and verified by investigators and clinicalresearch assistants (CRAs) at each study site. Theentered data will be systematically validated and auditedby the EDC system itself using edit checks, as well as byCRAs and data managers (DMs). Data query forms gen-erated by medically trained personnel will be sent to thesite investigators for identification. The electronic-casereport form (e-CRF) will be inspected by DMs at regularintervals throughout the trial to verify the protocol com-pliance. The entered data should be complete, consist-ent and accurate. The blinded immunology assay resultswill be sent in an additional Excel spreadsheet via emailto the data manager to be linked with the centraldatabase.

Statistical methodsData set: The full analysis set (FAS) will comply with theprinciples of intent-to-treat (ITT)21 and comprise allparticipants who receive at least one dose of vaccine andone follow-up visit. The per-protocol analysis set (PPS)will comprise ITT participants who have received theintegrated vaccination schedule and participated in allfollow-up visits. Whether a participant with protocoldeviations should be eliminated from the PPS set will bedetermined through blind data review process beforeanalysis. The safety analysis set will comprise participantswho were randomised to a certain group and vaccinatedincluding those with protocol deviations.Analysis principle: Analysis will be conducted following

the CONSORT22flow diagram (figure 1) of the phases

of the study: enrolment, allocation, interventions,follow-up and analysis. Baseline characteristics analysiswill be given for each group to depict the premises ofthis study using means, SD, minimum and maximumvalue as well as the two-sided 95% CIs (if any).Continuous data will be portrayed with means, SDs andCIs if a normal distribution hypothesis of data is met.Binomial data will be summarised as frequencies andproportions. All statistical tests will be demonstrated witha two-sided p value, and all CIs will be calculated as 95%at a two-sided level unless otherwise specified. A detailedstatistical analysis plan will be finalised before the firstsubstantive statistical analysis.Immunogenicity: The immunogenicity population con-

sists of all participants who belong to PPS and have avail-able serological data before and after immunisation.The primary immunogenicity end points for each of thepneumococcal serotypes are the proportion of partici-pants reaching the serotype-specific IgG concentrationthreshold of 0.35 µg/mL, and the geometric mean IgGantibody concentration, in the PCV13 group measured30 days after the primary series. Within each group, foreach pneumococcal serotype, the proportion of partici-pants with a specific-IgG antibody concentration thresh-old of 0.35 μg/mL and the two-sided 95% CI for theproportion will be calculated; for each antibody concen-tration, the value will be logarithmically transformed tobase 10, and the two-sided 95% CI will be structured. Toevaluate differences between two groups, the sequentialtest (figure 2) of hypotheses, including the non-inferiority for the seven common serotypes and the effi-cacy for the six additional serotypes, will be performed.The non-inferiority for each of the seven common sero-types will be declared if the lower limit of the two-sided97.5% (one-sided 98.75%) CI, which is calculated usingthe exact binomial method23 24 for the difference inproportions (PCV13—PCV7), is no <−10%, or the lowerbound of the two-sided 97.5% (one-sided 98.75%) CI ofthe IgG GMC ratio (PCV13/ PCV7) is >0.5. To controlfor familywise type I error, Bonferroni correction will beused to address multiple test, and thus a two-sided sig-nificance level of 0.025 (0.0125, one-sided) will beadopted. Only after each of the seven common serotypes

Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488 5

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

Figure 1 The study flow diagram.

Figure 2 The sequential test flow chart. GMC, geometric mean concentrations; PCV13, 13-valent pneumococcal conjugate

vaccine; PCV7, 7-valent pneumococcal conjugate vaccine.

6 Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

has been shown to be non-inferior will the other six add-itional serotypes be worth assessing. The efficacy criteriafor each of the six additional serotypes will be met ifboth the lower limit of the two-sided 95% CI for the pro-portion of vaccine recipients reaching the serotype-specific IgG concentration threshold of 0.35 μg/mLconducted by the normal approximation method is no<70%, and the two-sided 95% CI for the difference inproportion (PCV13–PCV7) compared by a χ2 test liesalmost to the right of the zero (zero or positive meandifference). A p value of <0.05 is considered statisticallysignificant.The secondary immunogenicity end points for each of

the pneumococcal serotypes in the PCV13 group are thepercentage of vaccine recipients reaching a serotype-specific IgG concentration threshold of 0.35 µg/mL30 days after the toddler dose and the percentages ofvaccine recipients reaching a serotype-specific IgG con-centration threshold of 1.0 µg/mL 30 days after theprimary series and booster dose. The proportions of theparticipants reaching the OPA titre threshold of 1:8 afterprimary and booster immunisation in predefined sub-groups also serve as the secondary immunogenicity endpoints. Within each group and for each antibody concen-tration or OPA titre, geometric mean will be calculatedand each concentration or titre will be logarithmicallytransformed for analysis. To assess differences betweenthe two groups and to test the non-inferiority of thePCV13 group, the analysis method described in theprimary end points section will be employed.As reported in previous studies,25 26 antibody

responses to serotypes 19A, 6A and 5 were noted in thePCV7 group. Therefore, when evaluating the efficacy of6A, 19A and 5 in the PCV13 group, the OPA resultsshould be combined.27

Safety: The safety population includes all participantsin the safety analysis set who will receive at least onedose of the study vaccine. Safety analysis including com-parisons of incidences of local reactions, systemic events,use of concomitant medications, dropouts and AEsbetween PCV13 and PCV7 will be demonstrated with thecorresponding two-sided 95% CIs and p values base onFisher’s exact test methodology for each dose, respect-ively, and for all in total as well. A p value equal to or<0.05 indicates a statistically significant difference.

HarmsAny SAEs will be reported to regulatory authorities,sponsors and Ethics Committees (ECs) within 24 hoursby fax, email or in person. Any suspected or unexpectedserious events will be reported to regulatory authoritiesby sponsors within seven calendar days and its follow-upinvestigation report will be submitted within eight calen-dar days of the first report submission day.

AuditingMonitoring of this trial will be conducted by a ContractResearch Organisation (CRO, Simoon Record Pharma

Information Consulting Co). Guidelines of GoodClinical Practice (GCP) will ensure compliance.

DISSEMINATIONInformed consentAll participants’ parents/legally responsible representa-tives will receive adequate information about the nature,purpose, and possible risks and benefits of the trial byan informed consent form approved by the EC. Theoriginal version of signed consent forms will be retainedby the investigators as a trial record, and a copy will bekept by the participants’ parents/legally responsiblerepresentatives.

ConfidentialityTo assure confidentiality, all participants will be allotteda unique subject number throughout the trial. Personalinformation will be collected with standardised paperforms by face-to-face interviews. Paper-based data will bestored in locked cabinets with limited access at each site,and the electronic data will be saved on password pro-tected computers.

Access to dataThe final data set will be under the custody of thesponsor together with the CRO selected beforehand.The data managers will be permitted to access the fullanalysed final data set. Access to the final or identifiabledata by others is subject to the sponsor’s approval.

Dissemination policyThe trial will be reported in accordance with theCONSORT guidance.

DiscussionUntil now, the efficacy and safety of PCV13 among theChinese population are still largely unknown. Hence,this study intends to characterise the PCV13 amongChinese infants and develop a safety profile for thevaccine. To fulfil these objectives, we calculated samplesizes under different hypothetical conditions and deter-mined the optimal sample size to use to take all factorsinto consideration. In study design, we adopted thesequential test method to inspect the non-inferiority ofeach of the seven common pneumococcal serotypes andthe efficacy of each of the six additional pneumococcalserotypes step by step to avoid false-positive error infla-tion. To be more discreet and conservative, instead ofsetting pneumococcal serotype 6B in PCV7 as a baselinewhen verifying the immunogenicity of each of the sixadditional serotypes in PCV13, as many other research-ers have done, we defined 70% as the lower limit of thetwo-sided 95% CI for the percentage of responders thatreach the serotype-specific IgG concentration thresholdof 0.35 μg/mL and supposed that the correspondingpercentages in PCV7 are <50%.29 Allowing for the cross-reactivation between 6A and 6B as well as 19A and 19B

Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488 7

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

in PCV7, we will select a subgroup of which to obtainthe OPA results for each of the 13 serotypes to correctfor potential bias.Given the main focus of the research, this study is not

designed to involve the assessment of concomitantimmunisation with other vaccines specified in paediatricvaccination schedules such as diphtheria, tetanus andacellular pertussis vaccine (DTaP), which might beregarded as a potential limitation. Since the feasibility ofPCV7 administered concomitantly with DTaP in main-land Chinese infants has been shown in the researchreported by Li RC et al,30 and the immunogenicity ofPCV13 coadministered with DTaP was demonstrated byTogashi T et al,7 we will not prohibit the usage of vac-cines listed in the expanded programme on immunisa-tion, but we will compulsively demand at least afortnight’s interval between two kinds of vaccination.

Author affiliations1Department of Health Statistics of Fourth Military Medical University, Xi’an,China2Walvax, Walvax Biotechnology Co., Ltd., Kunming, China3Beijing Chaoyang District Centre for Disease Control and Prevention, Beijing,China4Heibei Province Centre for Disease Control and Prevention, Shijiazhuang,China5Henan Province Centre for Disease Control and Prevention, Zhengzhou,China6Shanxi Province Centre for Disease Control and Prevention, Taiyuan, China7Guangxi Zhuang Autonomous Region Center for Disease Control andPrevention, Nanning, China

Contributors JLX provided statistical expertise and created the study designtogether with JJC. ZH and LY conceived the study and participated in studydesign. JJC drafted the manuscript and calculated the sample size. NMS, YLZ,SLX and GHL are the primary investigators and helped with revising theprotocol. RCL and YPL are the senior consultants and contributed to protocoldevelopment. SYY compiled the original protocol in Chinese and participated instudy design discussion. All authors read and approved the final manuscript.

Funding The study is supported by a grant from Bio-pharmaceutical MajorScientific and Technological Project (biological product development section)of Yunnan Province (project number 2015ZJ005). The funding source has norole in the trial design and will have no role in the trial conduct, data analysisand interpretation, or writing or reporting.

Disclaimer The views expressed are those of the author(s) and notnecessarily those of the DHSFMMU, Walvax or the Centre for Disease Controland Prevention.

Competing interests We have read and understood BMJ policy on declarationof interests and declare the following interests: This trial is sponsored byWalvax Biotechnology Co. Walvax, as the sponsor, is responsible fororganising and managing the study to support the prelicensing evaluation ofthe investigational PCV13. The new PCV13 was researched and developed byWalvax and manufactured by an affiliate of Walvax. JJC, LY, ZH and SYY workat Walvax and belong to the clinical research team of PCV13.

Ethics approval The trial is approved by the Beijing Chaoyang District Centrefor Disease Control and Prevention Ethics Committee (the lead institution), theHeibei Province Centre for Disease Control and Prevention Ethics Committee,the Henan Province Centre for Disease Control and Prevention EthicsCommittee, and the Shanxi Province Centre for Disease Control and PreventionEthics Committee. The registration number of this trial is NCT02736240.

Provenance and peer review Not commissioned; externally peer reviewed.

Open Access This is an Open Access article distributed in accordance withthe Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license,

which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, providedthe original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/

REFERENCES1. WHO. Pneumococcal conjugate vaccine for childhood immunization

—WHO position paper. Wkly Epidemiol Rec 2007;82:93–104.2. O’Brien KL, Wolfson LJ, Watt JP, et al. Burden of disease caused by

Streptococcus pneumoniae in children younger than 5 years: globalestimates. Lancet 2009;374:893–902.

3. WHO. Pneumonia WHO Fact sheet N°331. http://www.who.int/mediacentre/factsheets/fs331/en/ (accessed 25 Nov 2015).

4. American Academy of Pediatrics Committee on Infectious D.Recommendations for the prevention of Streptococcus pneumoniaeinfections in infants and children: use of 13-valent pneumococcalconjugate vaccine (PCV13) and pneumococcal polysaccharidevaccine (PPSV23). Pediatrics 2010;126:186–90.

5. Black S. The volatile nature of pneumococcal serotypeepidemiology: potential for misinterpretation. Pediatr Infect Dis J2010;29:301–3.

6. Yeh SH, Gurtman A, Hurley DC, et al. Immunogenicity and safety of13-valent pneumococcal conjugate vaccine in infants and toddlers.Pediatrics 2010;126:e493–505.

7. Togashi T, Okada K, Yamaji M, et al. Immunogenicity and safety ofa 13-valent pneumococcal conjugate vaccine given with DTaPvaccine in healthy infants in Japan. Pediatr Infect Dis J2015;34:1096–104.

8. Spijkerman J, Veenhoven RH, Wijmenga-Monsuur AJ, et al.Immunogenicity of 13-valent pneumococcal conjugate vaccineadministered according to 4 different primary immunizationschedules in infants: a randomized clinical trial. JAMA2013;310:930–7.

9. Deloria Knoll M, Park DE, Johnson TS, et al. Systematic review ofthe effect of pneumococcal conjugate vaccine dosing schedules onimmunogenicity. Pediatr Infect Dis J 2014;33(Suppl 2):S119–29.

10. WHO. Pneumococcal vaccines WHO position paper—2012. WklyEpidemiol Rec 2012:129–44.

11. Mitchell R, Trück J, Pollard AJ. Use of the 13-valent pneumococcalconjugate vaccine in children and adolescents aged 6–17 years.Expert Opin Biol Ther 2013;13:1451–65.

12. De Montalembert M, Abboud MR, Fiquet A, et al. 13-valentpneumococcal conjugate vaccine (PCV13) is immunogenic and safein children 6-17 years of age with sickle cell disease previouslyvaccinated with 23-valent pneumococcal polysaccharide vaccine(PPSV23): results of a phase 3 study. Pediatr Blood Cancer2015;62:1427–36.

13. Bryant KA, Frenck R, Gurtman A, et al. Immunogenicity and safetyof a 13-valent pneumococcal conjugate vaccine in adults 18-49years of age, naive to 23-valent pneumococcal polysaccharidevaccine. Vaccine 2015;33:5854–60.

14. Durando P, Rosselli R, Cremonesi I, et al. Safety and tolerability of13-valent pneumococcal conjugate vaccine in the elderly. HumVaccin Immunother 2015;11:172–7.

15. Jackson LA, Gurtman A, Rice K, et al. Immunogenicity and safety ofa 13-valent pneumococcal conjugate vaccine in adults 70 years ofage and older previously vaccinated with 23-valent pneumococcalpolysaccharide vaccine. Vaccine 2013;31:3585–93.

16. Greenberg RN, Gurtman A, Frenck RW, et al. Sequentialadministration of 13-valent pneumococcal conjugate vaccineand 23-valent pneumococcal polysaccharide vaccine inpneumococcal vaccine-naive adults 60-64 years of age. Vaccine2014;32:2364–74.

17. WHO. Guidelines on clinical evaluation of vaccines: regulatoryexpectations. WHO Tech Rep Ser 2004;924:55–92.

18. Wernette CM, Frasch CE, Madore D, et al. Enzyme-linkedimmunosorbent assay for quantitation of human antibodies topneumococcal polysaccharides. Clin Vaccine Immunol2003;10:514–19.

19. Siber GR, Chang I, Baker S, et al. Estimating the protectiveconcentration of anti-pneumococcal capsular polysaccharideantibodies. Vaccine 2007;25:3816–26.

20. Leach AJ, Mulholland EK, Santosham M, et al. Pneumococcalconjugate vaccines PREVenar13 and SynflorIX in sequence oralone in high-risk Indigenous infants (PREV-IX_COMBO): protocolof a randomised controlled trial. BMJ Open 2015;5:e007247.

21. ICH Efficacy Guidelines topic E9 guideline for good clinical practice,step 4 version. 1998. http://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines (accessed 10 Sep 2015).

8 Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from

22. Schulz KF, Altman DG, Moher D, et al. CONSORT 2010 statement:updated guidelines for reporting parallel group randomised trials. IntJ Surg 2011;9:672–7.

23. NEWCOMBE RG. Interval estimation for the difference betweenindependent proportions comparison of eleven methods. Stat Med1998;17:873–90.

24. Newcombe RG. Two-sided confidence intervals for the singleproportion comparison of seven methods. Stat Med1998;17:857–72.

25. Esposito S, Tansey S, Thompson A, et al. Safety andimmunogenicity of a 13-valent pneumococcal conjugate vaccinecompared to those of a 7-valent pneumococcal conjugate vaccinegiven as a three-dose series with routine vaccines in healthy infantsand toddlers. Clin Vaccine Immunol 2010;17:1017–26.

26. Olivier C, Belohradsky BH, Stojanov S, et al. Immunogenicity,reactogenicity, and safety of a seven-valent pneumococcal

conjugate vaccine (PCV7) concurrently administered with a fullyliquid DTPa-IPV-HBV-Hib combination vaccine in healthy infants.Vaccine 2008;26:3142–52.

27. Cooper D, Yu X, Sidhu M, et al. The 13-valent pneumococcalconjugate vaccine (PCV13) elicits cross-functionalopsonophagocytic killing responses in humans to Streptococcuspneumoniae serotypes 6C and 7A. Vaccine 2011;29:7207–11.

28. ICH Efficacy Guidelines topic E6 guideline for good clinical practice,step 4 version. 1996. http://www.ich.org/products/guidelines/efficacy/article/efficacy-guidelines.html (accessed 5 Aug 2015).

29. Elberse KE, de Greeff SC, Wattimena N, et al. Seroprevalence ofIgG antibodies against 13 vaccine Streptococcus pneumoniaeserotypes in the Netherlands. Vaccine 2011;29:1029–35.

30. Li RC, Li FX, Li YP, et al. Safety and immunogenicity of a 7-valentpneumococcal conjugate vaccine (Prevenar): primary dosing seriesin healthy Chinese infants. Vaccine 2008;26:2260–9.

Chen JJ, et al. BMJ Open 2016;6:e012488. doi:10.1136/bmjopen-2016-012488 9

Open Access

on February 1, 2020 by guest. P

rotected by copyright.http://bm

jopen.bmj.com

/B

MJ O

pen: first published as 10.1136/bmjopen-2016-012488 on 19 O

ctober 2016. Dow

nloaded from